中国农业气象 ›› 2013, Vol. 34 ›› Issue (03): 342-349.doi: 10.3969/j.issn.1000-6362.2013.03.015

• 论文 • 上一篇    下一篇

北方地区日光温室气象灾害风险评价

杨再强1，2，张婷华1，黄海静3，朱凯1，张波1   

  1. 1南京信息工程大学江苏省农业气象重点实验室，南京210044；2南京信息工程大学应用气象学院，南京210044；3海南省气候中心，海口570203
  • 收稿日期:2012-08-24 出版日期:2013-06-20 发布日期:2013-06-17
  • 作者简介:杨再强（1967-），四川安岳人，博士，教授，博士生导师，主要从事设施农业气象灾害预警研究。 Email： yzq@nuist.edu.cn
  • 基金资助:
    公益性行业（气象）科研专项（GYHY(QX) 201206024；GYHY(QX) 201006028）；江苏省科技支撑项目（社会发展，BE2010734）

Meteorological Disaster Risk Evaluation of Solar Greenhouse in North China

YANG Zai qiang1,2,ZHANG Ting hua1,HUANG Hai jing3,ZHU Kai1,ZHANG Bo1   

  1. 1Jiangsu Key Laboratory of Agricultural Meteorology,Nanjing University of Information Science & Technology,Nanjing210044,China;2College of Applied Meteorology,Nanjing University of Information Science & Technology,Nanjing210044;3Hainan Climate Center, Haikou570203
  • Received:2012-08-24 Online:2013-06-20 Published:2013-06-17

PDF (PC)

3552

被引次数

37

摘要： 利用北方典型日光温室室内以及相应台站气象数据，建立基于BP神经网络的室内气温预报模型，以此预报北方243个台站1990-2009年室内气温，利用预报的室内气温数据及室外降水、日照、风速等气象数据和主要气象灾害指标，构建基于实数编码的加速遗传算法（Real code accelerating genetic algorithm，RAGA)和投影寻踪（Projection pursuit evaluate，PPE)的日光温室气象灾害风险评价模型，并对北方地区日光温室主要生产月气象灾害风险进行逐月评价。结果表明，北方日光温室室内气温预报值与实际观测值的标准误差在0.89～1.54℃，方程决定系数在0.87～0.94。北方地区日光温室1-3月的气象灾害风险等级较高，分布在天山以北、大兴安岭以北的地区和西藏地区，主要是低温和大风沙尘天气。9月气象灾害风险等级最低，分布在长城以南，黄河以北地区，主要是低温。本研究构建的北方地区日光温室气象灾害风险评价模型可为日光温室气象灾害风险区划和防御提供决策支持。

关键词: RAGA, BP神经网络, 投影寻踪, 日光温室, 风险评价

Abstract: The temperature prediction model indoor based on BP neural network was established,based on meteorological data inside typical solar greenhouse in North China and other meteorological stations.The temperature indoor of 243 meteorological stations was forecasted by the simulation model.Comprehensive meteorological risk assessment model for solar greenhouse was established based on real code accelerating genetic algorithm (RAGA) and projection pursuit evaluate model (PPE),with forecast temperature data indoor and precipitation,sunlight and wind speed from other meteorological stations.The meteorological disaster risk of solar greenhouse in North China was evaluated.The results showed that the standard error was 0.89-1.54℃ between forecasted temperature and observed data,and the determination coefficient was between 0.87-0.94.The highest meteorological disasters risk level of solar greenhouse was from January to March,which was located in North of Tianshan,North of Xing Anling and Tibetan,mainly because of low temperature and frequent winds dust.The lowest risk level of meteorological disasters was in September,which was located between south of the Great Wall and north of the Yellow River,mainly because of low temperature.The meteorological risk evaluation model could provide decision making support for distribution and defence of agro meteorological disaster risk.

Key words: RAGA, BP neural network, Projection pursuit, Solar greenhouse, Risk evaluation

版权所有 © 《中国农业气象》编辑部
地址：北京中关村南大街12号中国农科院环发所　邮编：100081
电话：010-82109774　 Email：leishuiling@caas.cn; zgnyqx@caas.cn
本系统由北京玛格泰克科技发展有限公司设计开发